Is accelerating down hills fuel efficient?

[EcoCivic]

I have always accelerated when going down hills to maximize the effect of gravity, but I wonder if I am doing the right thing for mileage now. What do you Ecomodding pros do on hills? Thanks, I am pretty new to hypermiling.

[JockoT]

If there is a climb ahead I will gently accelerate down a hill, to build up some momentum for the climb ahead. A bit of light throttle there, uses a lot less fuel than starting the hill at a lower speed and having to burn fuel to get up it.
If there is no climb ahead I would probably coast down the hill, or take advantage of DFCO. The last thing you want to do is accelerate down the hill, then have to brake because of a bend or other hazard to be negotiated.

[slowmover]

With my four-ton pickup (and more so with 35' travel trailer in tow) I enter a highway with the intention of of using terrain to help the last 10-15/mph up to set cruising speed.

That is, I'll accelerate up past the 45-mph legal minimum, but I wish to use any sort of downgrade to get it up to the usual 59-mph and set the cruise control.

Terrain and traffic don't always cooperate in this, but it does lessen the engine load when those other factors allow for it.

I prefer that commercial traffic be a quarter-mile back, and cars less so.

Alternately,

One uses the brakes, not the throttle, to merge onto an Interstate. Ideally one is at or above the legal maximum well before the end of the ramp (most entrances are downhill for this reason). In this use of a downgrade the penalty is minimal compared to the safety advantage.

Choosing the point of merge means more in my experience.

As to rolling terrain I find that setting cruise control back by two mph obviates the penalty of its use.

With hills, I let it run out some. Maybe five mph downgrade. On an upgrade, more of a change in speed is acceptable. But not to the point of having traffic jam behind me (relative to flatland steady-state).

As I run 10k miles monthly as a pro driver, spacing with other vehicles is paramount. I can't do a thing about tailgaters (less than 100' behind me at speed), but there's no way I'll allow things to jam up ahead of me (less than 200').

So, yeah, aceelerate when there's less penalty, and don't exceed 80% Engine Load in a climb.

Frankly, we most of us run so few highway miles annually (this based on engine hours) that what we do out there past a sensible set speed (60 is The Wall) doesn't mean much to the calendar average.

IOW, with no lane changes and no use of brakes, 58 or 62 or 66 don't mean much. It's tire wear and other factors pertaining to longevity that factor higher.

.

[JockoT]

Here in Scotland most on ramps are uphill. Add to that short slip roads and trucks struggle.

[oldtamiyaphile]

People confuse 'speed' with 'energy'.

Unless the hill is so steep that you have to drop back gears or go into fuel enrichment, gaining momentum has the same cost on the flat, up hill and down hill. What goes up, must come down, essentially.

Ideally, you want to crest the hill at such a speed that you end up at the speed limit (or your target speed) at the bottom.

Only accelerate on descents if the next hill you face requires that extra momentum. Otherwise, climb slow and steady, in top gear at BSFC (80% load).

*All assuming you have a NA, throttled spark ignition engine.

I've run a test on hills and indeed you get better economy with hills than without. I have a great hill that I use regularly. It's a steep climb, followed by a shallow decent. BSFC up, EOC down. Crazy mileage.

[Ecky]

"It depends". As a general rule, I'll engine-off coast down hills and keep high load up hills - unless the hill is steep enough that I'd need to downshift, in which case I will accelerate (partway) down the hill to make sure I can get up the next one.

[D15r]

In simple terms It's always more efficient to accelerate downhill than it is on the flat or up hill. But it needs to be applied with situational context and that's the tricky part which the driver has to compute :-P

[Panther140]

Originally Posted by D15r

In simple terms It's always more efficient to accelerate downhill than it is on the flat or up hill. But it needs to be applied with situational context and that's the tricky part which the driver has to compute :-P

Its actually the other way around.

If you use your engine to accelerate down hills until the vehicle has XXX amount of energy, you will encounter higher wind speed than if you accelerate up a hill to get that same amount of energy.

That is because when you accelerate down hills, all of the energy is momentum.

When you accelerate up hills, lets say half of that energy is potential energy.

With some of the energy being in the form of potential energy, you don't encounter the same drastic wind drag losses that you would encounter if all of the energy were in the form of momentum (aka, high speeds).

Wind drag increases exponentially as speed increase, and it becomes quite costly as you go above about 60mph.

I calculated it once. It takes multiple times more energy to overcome wind drag at 80 than it does to overcome wind drag at 55.

So if you accelerate up a big hill, but hit a peak speed of 55, and then coast down the other side for 3 miles, one gallon of gas goes farther than if you spent that gallon of gas accelerating down the hill and plowing wind at 80mph.

Coast down one hill, you'll coast down to 45mph by the next hill. then accelerate up that hill, coast down the other side etc etc.

If you accelerate down that hill, you have to also overcome wind drag and still burn a lot of gas climbing the next hill.

[oldtamiyaphile]

Originally Posted by D15r

In simple terms It's always more efficient to accelerate downhill than it is on the flat or up hill. But it needs to be applied with situational context and that's the tricky part which the driver has to compute :-P

More economical, not more efficient.

[D15r]

Your adding in a lot of varibles, who said anything about what speeds are involved? It takes more fuel to accelerate up hill than it does down hill, that can't be argued. Adding varibles it can be argued as you have done so.