Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

Hypermiling techniques?

This site may earn commission on affiliate links.
Aside from going 25 mph and only downhill, has anyone figured out the most efficient driving technique for the Model S? In particular when getting to highway speeds, but in general as well. I've tried slow acceleration to highway speeds, a blast of power, a mix, and just can't tell what is best. Same for surface roads. I'm pretty good at being efficient below 40mph, but then what is the point of having such a powerful car?

Thanks and sorry if I duplicated a thread.
 
I thought it interesting to see just how gradually a resume on the cruise control speeds up. I leap to the assumption that this acceleration (or lack thereof :) is more efficient than blasting up to the desired speed.

Many freeway on-ramps here in SoCal are uphill. I find that trying to accelerate up the hill appears to use more energy than if I just trickle up the hill and then accelerate when the onramp levels off with the freeway.

I can say that the Model S gets great range in SoCal bumper to bumper traffic. People are a lot more tolerant of slowly accelerating to 15 or 20mph than they are at a stop light. Keeping pace and distance to allow you to coast longer rather than stop and go you can get low 200's watt/mi.
 
Many freeway on-ramps here in SoCal are uphill. I find that trying to accelerate up the hill appears to use more energy than if I just trickle up the hill and then accelerate when the onramp levels off with the freeway.

I can say that the Model S gets great range in SoCal bumper to bumper traffic. People are a lot more tolerant of slowly accelerating to 15 or 20mph than they are at a stop light. Keeping pace and distance to allow you to coast longer rather than stop and go you can get low 200's watt/mi.

What you have noted is true for all EV, etc. My Volt loves slow speed. Same is true for the Leaf and MS.
 
Get behind a truck. You don't have to tailgate at all (it's dangerous, anyway), but just being being an 18-wheeler, or any tall truck, reduces wind resistance tremendously. I learned this the first week I had the car when I was charging with a 110v outlet and had to conserve every Wh. There was nothing better for conserving energy than getting behind a truck on the highway (at any speed) and letting the truck take the brunt of the wind resistance. The only downside is you are generally going to go slower than you otherwise would because trucks drive slower than cars, on average, and you don't want a truck throwing off debris and hitting your car!

I'm convinced that someone who wants to hire a truck to drive at 50 mph in front of his could hit the 400 miles/charge challenge from Elon. I don't have the time or inclination to try this, but it definitely fits the bill.
 
Get behind a truck. You don't have to tailgate at all (it's dangerous, anyway), but just being being an 18-wheeler, or any tall truck, reduces wind resistance tremendously.

+1. On a recent trip the temperature was -15C (5F), which would normally mean a hit on range of about 20%. I got behind a tandem tanker truck going 107 kph (67 mph) as an experiment. I was very surprised to discover that I could get "Rated Range" at that speed and temperature. Quite impressive!
 
Aside from going 25 mph and only downhill, has anyone figured out the most efficient driving technique for the Model S? In particular when getting to highway speeds, but in general as well. I've tried slow acceleration to highway speeds, a blast of power, a mix, and just can't tell what is best. Same for surface roads. I'm pretty good at being efficient below 40mph, but then what is the point of having such a powerful car?

Thanks and sorry if I duplicated a thread.

I think it's well established that hard acceleration is inefficient. Certainly, you should seek to avoid using the friction brakes as much as possible. You should also avoid using regen, since it only recaptures a portion of the energy required to accelerate the car; on the flat, that means lifting off the accelerator just enough to reduce current draw to zero, which has to happen much earlier than you are used to when approaching an intersection. You'll want to use regen to the minimum extent possible to slow the car as you approach a stop.

The discussion on the thread linked by smorgasbord considers the question of whether to use regen when descending a hill or to just let the car speed up in neutral, though I did talk about generic hypermiling techniques in this post on that thread. The emerging consensus is that using regen to limit your speed on a downhill is a good way to recapture some of the potential energy of the descent that you would otherwise lose through aero drag at higher speeds.
 
I think it's well established that hard acceleration is inefficient. Certainly, you should seek to avoid using the friction brakes as much as possible. You should also avoid using regen, since it only recaptures a portion of the energy required to accelerate the car; on the flat, that means lifting off the accelerator just enough to reduce current draw to zero, which has to happen much earlier than you are used to when approaching an intersection. You'll want to use regen to the minimum extent possible to slow the car as you approach a stop.

I'm still wondering about this one. I have questions about both halves of the above statements. And yes, I know it's an old thread.

Isn't the energy usage the area under the curve? Therefore, would it be more efficient to rapidly get up to the desired speed, then lift up the accelerator pedal and run for a longer period of time using a smaller amount of energy vs slowly accelerating up to speed and staying there? In the first scenario, you use far more energy per second, but for a far shorter period of time.

In regards to stopping, if you can't anticipate a light turning red, doesn't it pay to feather the accelerator to maintain regeneration for the longest time possible, even at lesser amounts of regeneration vs keeping the car rolling as long as possible, then taking your foot off the accelerator and getting maximal regeneration, but for a far shorter period of time (perhaps 1 second at most compared to 6-10 seconds by feathering the accelerator?
 
Last edited:
I think it's well established that hard acceleration is inefficient. Certainly, you should seek to avoid using the friction brakes as much as possible. You should also avoid using regen, since it only recaptures a portion of the energy required to accelerate the car; on the flat, that means lifting off the accelerator just enough to reduce current draw to zero, which has to happen much earlier than you are used to when approaching an intersection. You'll want to use regen to the minimum extent possible to slow the car as you approach a stop.

QUOTE]

I'm still wondering about this one. I have questions about both halves of the above statements. And yes, I know it's an old thread.

Isn't the energy usage the area under the curve? Therefore, would it be more efficient to rapidly get up to the desired speed, then lift up the accelerator pedal and run for a longer period of time using a smaller amount of energy vs slowly accelerating up to speed and staying there? In the first scenario, you use far more energy per second, but for a far shorter period of time.

In regards to stopping, if you can't anticipate a light turning red, doesn't it pay to feather the accelerator to maintain regeneration for the longest time possible, even at lesser amounts of regeneration vs keeping the car rolling as long as possible, then taking your foot off the accelerator and getting maximal regeneration, but for a far shorter period of time (perhaps 1 second at most compared to 6-10 seconds by feathering the accelerator?

Acceleration: if a single-motor drivetrain you have a simple efficiency curve which says that it's better to use less power. Also, basically, the power required to maintain speed increases more rapidly than speed. As a result slower is intuitively better. But, there's a fly in the ointment: non-drivetrain power, which includes the battery heating/cooling, cabin climate control, lighting and computer systems.

Let x be extra power of non-drivetrain use. Let s be speed in mph. Let p be power to maintain speed s. Let a be additional power to maintain speed i mph higher.
Time saved per hour = ni/n(n+i) = i/(n+i). If (1-i/(n+i))(p+a+x) < p+x, it's better to use the extra power, but given that time savings/kW will rapidly diminish with speed (as n increases, for a given a, i decreases), I suspect that the speed and acceleration windows won't be very large.

Braking:
As I understand it, yes, the general rule is to maintain steady, low regen, over none and then a lot. Decide on a point of no return in advance, and then be ready to react to the change to yellow., so you can regen low and steadily if the light changes. But, consider other drivers: when you decelerate early you reduce the available space on the road for other drivers so you may contribute to congestion behind you. Also, if approaching an I tersection with a turning lane, remember that drivers behind you may miss a cycle if they can't turn early.
 
Last edited:
... would it be more efficient to rapidly get up to the desired speed, then lift up the accelerator pedal and run for a longer period of time using a smaller amount of energy vs slowly accelerating up to speed and staying there? In the first scenario, you use far more energy per second, but for a far shorter period of time.

Our mind fools us here. We think getting something bad over with quickly and spending more time on the good side would be better. But it's only our perception. Numbers don't support it. Hard acceleration causes higher losses in the battery, inverter/motor and other places. As a result you need more energy to reach your desired speed than doing it at a slower rate. The amount of energy needed to accelerate to a given speed is the exact same no matter how slow or quickly you do it. So doing it quickly doesn't help at all. But the amount of losses are higher in quick acceleration in an EV.

Same is true for slowing the car down and using regen. It is more efficient to do it gently over the entire distance to your final stop than doing it hard at the end. Same as before, the higher the power the higher the losses. Ohmic losses (losses dues to resistance in an electric circuit) are geometric, meaning at double the power the losses are 4 times. At 4 times the power the losses are 8 times.

Regen is definitely better going downhill than letting the car speed up to a point where wind drag limits it's speed. Imagine you go down for 5 miles. You can get a little back going 45 using regen. Or you can let it roll free at 70 mph and get nothing back into the battery. At the end of the 5 miles you will have used zero energy going 70, vs having gained back maybe 1 or 2 kWh using regen. Of course you went slower using regen. And that's where it comes down to the first rule of hypermiling: go slow.

BTW, good point from the previous poster about the constant load (AC, heater). Those contribute more the longer your drive is. The good thing about these, though is that they are optional. You can turn them off to eliminate them for hypermileing.
 
Last edited:
  • Like
Reactions: hybridbear
Car A
Accelerate 0-60 in 4.4s, finish the 1/4 mile.

Car B
Accelerate 0-60 in 5.6s, finish the 1/4 mile.


1. Time range (4.4, 5.6]s
A will be travelling at 60mph while B will be travelling at < 60mph. A will be spending more energy to counteract higher wind resistance.

2. Time range (4.3, 4.4]s
A will be travelling near 60mph whereas B will likely be closer to 50mph. A will be spending more energy to counteract higher wind resistance.

3. Time ranges (0.0, 0.1] through (4.2, 4.3]
A spends more on wind resistance, again.

4. Time range [5.6, {end-of-quarter-mile})
Here it's a little trickier because A has less distance to travel than B at this point. I'm too lazy to do the math on this one (right now at least) but I presume the deficit from 1, 2, 3 are too big to overcome to make A the more conservative power consumer.


My point being, I think on wind resistance alone you can show slower acceleration is cheaper.
 
Rate of acceleration in my P85 has a significant impact on my Wh/mi numbers. When I keep acceleration under 40 kW, my efficiency goes way up. If I can manage to accelerate no higher than 20 kW, which is challenging, I can achieve Wh/mi numbers in the 250s.
 
Rate of acceleration in my P85 has a significant impact on my Wh/mi numbers. When I keep acceleration under 40 kW, my efficiency goes way up. If I can manage to accelerate no higher than 20 kW, which is challenging, I can achieve Wh/mi numbers in the 250s.

That's been my experience as well. So far i've achieved a "lifetime" average of 310 wh/mi over 45,000 miles (i took possession in Dec. of 2012). Living in the Northeast we get seriously taxed by these harsh winters - this past summer, on a road trip to North Carolina in mild temps i had an average of 275 wh/mi for 1,500 miles, and that's with highway speeds of 70-70mph.

Drafting off trucks used to help the MINI E i had, but the Model S is so aerodynamic that i haven't noticed that big a bonus with it. I used to throw it in neutral on downhills, until the technician mentioned that he was going to have to change out some air vents that weren't opening as designed - when i confessed to the neutral-throttling, he said that's why and it's not a good idea, so i stopped. I now try and limit the regen by feathering to a true neutral or low-regen state, and find that gives me the best results. It's lots of fun working all the angles - sure makes long drives go by faster.
 
Yes it is. If you get up to maximum speed quickly, there is more area under the curve because it's "box" shaped rather than sloped. Therefore more energy is used.
But the width of the box is less, so the area is the same. E=1/2 Mv^2 - it doesn't matter from a pure energy standpoint how fast or slow you get there. In the S, getting there fast is worse because of increased losses due to higher current causing more inefficient power transfer.
 
But the width of the box is less, so the area is the same. E=1/2 Mv^2 - it doesn't matter from a pure energy standpoint how fast or slow you get there. In the S, getting there fast is worse because of increased losses due to higher current causing more inefficient power transfer.

No, the X is the distance traveled--not the time it takes.

- - - Updated - - -

What's the explanation behind this?
Presumably the algorithm doesn't open the vents when in neutral. Why they programmed it that way, no idea. Maybe it was easier, or maybe it's for diagnostic purposes, as in they run some diagnostic tests and the vents remain closed, or the vents open and close with a certain diagnostic routine.