It's pretty astonishing how much greater knowledge of my car's behaviour changes even small driving decisions. I know the novelty will wear off, but it's making it a lot more interesting, and I hate driving so that's saying something.

That's one of the things I like about hypermiling...it gives me something to do. And, even better, it actually increases the attention I pay to the road and traffic in the process of occupying my mind!

Some people hyper-mile scientifically, with a lot of hard data and mechanical inputs. They try to become a part of the machine. Maybe they do not understand the real physics of the act, but they accomplish the same results.

Others who are more intimately familiar with the machine and the way you interact with that machine are more of an artist at hypermiling. I like to think I fall into the second category.

My father once flew a gal from Alabama to Virginia in a AT6 Texan (primary trainer), right after he came back from WW2 and his 30 missions in a B17 over Europe.

When they landed the woman asked him how he controlled the airplane. He showed here the stick and rudder pedals and how they moved the control surfaces. She responded. I watched them for 4 hours and never saw either or them move. His control inputs were so small, they were almost invisible.

He took pride in returning home with more fuel in the tanks than any other plane in his squadron. Fuel meant life, when you could be damaged and loose that precious reserve.

I try to drive much the same way he flew that plane, concentrating on maintaining my inertial state and anticipating any future necessity to change that inertial state due to other drivers, traffic signals or any other situation that would rob me of my hard earned inertia.

I have noticed that with the Insight which gives me instant fuel consumption readings that some of my pulses are very gradual. The rationale is if it takes you longer in the pulse but uses less fuel in any instant, that real question is how do you best get to your peak pulse speed with the least fuel, but you also have to consider the fact that a more gradual pulse covers a greater distance and may be more economical in the long run.

Every car or truck is different in it's specific requirements. Every engine has its preferences.

Most here have read about my point of maximum effective compression in any engine being the most efficient state of operation. The work any engine produces is a form of leverage with the difference between the compression ratio and the expansion ratio being the lever. Minimum throttle in highest gear gives you that state of operation. It does not require any scientific instrumentation to achieve that state of operation. When you press down on the throttle more nothing happens, unless you downshift.

If you need a greater rate of acceleration then you must downshift. In all other situations use the highest practical gear. I use 5th at 30 MPH, even when the engine is cold. That's lot lower speed that the manual recommends, but I am not trying to accelerate, just maintaining a steady speed. This also helps the car to warm up faster with less fuel.

I can not tell you it will work for you, but I can guarantee you it works for me.
Using the very slightly graded hills here I can P&G without people around me even knowing I am doing anything unusual. I do find that some other drivers will tailgate me no matter what speed I am going. When I see them doing that I will extend my glide until they decide to pass me. I don't want them around me because they are not really paying attention to what they are doing.

Every driving environment is different. There is no set rules about how you adapt to your environment. When you have learned P&G in the mechanical sense, take it to the next step which is to make it an art, and ingrain the method into your driving style until it becomes second nature.

When you do this you will see a very consistent result (look at my Insight gas logs), with very small differences in mileage from tank to tank. That will only change when your driving environment changes significantly.

regards
gary

Yes! Now that my responses are more influenced by cars around me my attention is definitely greater. Admittedly instead of "will this guy brake unexpectedly" I'm thinking "will this SOB make me waste energy braking when we're well under the speed limit anyway; moron!", but the effect is the same.

There are some who view engineering as the junction of art and science; that's what I aspire to. Admittedly, in practice it's the junction because it destroys one by turning it into the other. :)

Now Marvin definitely likes brisk acceleration, you've gotta get the RPM over 2000 or he just splashes his ATF around a lot. He is relatively easy to drive in "linear momentum hybrid" mode.

IMO the advice about acceleration, should be don't accelerate too long, or too much, rather than too hard, too long means you have to use the brakes, or "Don't attain a speed higher than that necessary to roll out to the next stoplight"

Wile-E is a PITA, various techniques barely seem to have an effect. But I just revamped the PCV system and he drives different again, so maybe that will help. He's hard to "LMH" I think he uses just as much fuel idling as driving along on the roads where I can try it. Don't really do it on the highway, because traffic is doing 75 or 80 in a 60 limit, I'm driving around 60-65, 20mph slower than most of the rest of the traffic, I could technically drop to 45-50 without legally being considered an obstruction.... but the speed differential would be getting suicidal. Actually it's a bit easier in Marvin, he gets more "road respect" as it were.

Just ran a program to determine the best combination of throttle, min speed, and max speed based on SG2 data I collected. I smoothed the curves out a bit, but of course there'll still be noise. Since a four dimensional graph is challenging to represent, I simply sorted the results by mpg.

Not too surprisingly, a cycle with min and max both very low yields the best economy, but I'd never be willing to employ such a labor intensive approach. The worst is also a small cycle, but with min and max at a higher speed and with the lowest throttle, so the most drag losses are encountered and the engine not run at an efficient load. Ironically, a good compromise of distance traveled per cycle and economy are the settings I initially tried just by chance: 25 to 60 mph, accelerating with a throttle of 25. That yields a theoretical 65 mpg and covers about .95 of a mile per P&G cycle. In practice, things are seldom flat here and my behaviour is controlled more by hills, speed limits, and traffic. In winter I won't P&G at all because of the risk associated with higher speeds on the frequently sloppy roads.

Next I want to record a coast down from high speed with the throttle held at 10 (yields a 35 mph level travel speed). I can then rerun my program to determine how well simply pulsing the throttle could work. Not as efficient as going into neutral, but a lot less effort so worth some exploration.

Do you mean just on and off the throttle? I call that Pulse & DFCO, and it increased my fuel usage vs. steady throttle, but your car is different and you will certainly do it differently than me. I wouldn't recommend it as a strategy in a production environment but go ahead and test it to measure how it works for you.

No, I meant high throttle alternating with a lower one. That way the engine's losses are still justified because you're still producing power at a useful rate, but you shed the high speed from the pulse (that's the theory anyway). But since you bring it up, I may as well gather DFCO coast down info while I'm at my test track and figure that out too. Why not?

Yeah, get all the data you can.

The issue with Pulse & DFCO should still affect Pulse & Smaller Pulse (or whatever you'd like to call it), if I understand correctly what's going on. I think the problem is that during the DFCO or smaller pulse, it's just wasting too much energy on overhead -- engine friction, reciprocating loss, accessories, etc. For every revolution*, you've got the same reciprocating loss, the same accessory load, and most of the same engine friction (I think it goes up some with load).

*: Note that I said revolution, not RPM; you have the same quantity of revolutions in the same distance if you're in the same gear.

Yep, that's the general "big engine, small load" problem. I got thinking along these lines when I discovered the best steady state speed for a level with my car was 35 mph (TPS = 10). I realize that low drag is a big part of that, but figure if the overhead were too ridiculous at that power level it'd swamp drag savings (and was what I expected to see when I started gathering data!). Should be interesting.