Google Earth info
 

I was just looking on google earth for elevations of my ride to check out all of my to work options. I didnt find anything better than my current one however I did find that my one EOC goes from an elevation of 1903 down to 799 over about 6 miles. Kind of funny because if I was looking over a 1200 foot cliff I would be sick but drive that everyday.

Cool! But the return drive must SUCK going back up 1200 ft.

Ya it kind of does but I like it alot better knowing I came down "for free"

This brings up a question I've had before. Does going up-hill cause more of a penalty than the gain you get from going down the same hill? Perhaps this is a good item for the experiments section? If so, how would you go about testing it?

It would be tough to test on city "courses", but I think if enough runs were performed then you could at least deduce the benefits after figuring the margin of error between runs.

First you'd have to do a number of "normal driving" runs with no FE technique. Once you have a sufficient baseline established then you could try some FE technique runs and see how much gain you acheive down and up hill.

Absolutely (in my mind). Anytime you transfer a form of energy, you have to figure in some kinds of loss. When you're going down a hill, you're transfering potential energy to kinetic energy, but you're also losing energy to drag and to tire friction (among other things). When you try to go back up the hill, you're still losing energy to the same parasitic losses.

It works great if you get to coast down the hill at no fuel use and without braking to loose energy. Going up hill runs the engine at a higher load but not quite half the MPG of level driving and going down hill returns a much higher MPG generally.

Just setup the SGI in a Dodge Caravan V6 3.3 liter - the thing suck 0.5gph at idle!!!! My xB used 0.2-0.1gph for 1.5liters - holy cow!